Massey Documents by Type

Permanent URI for this communityhttps://mro.massey.ac.nz/handle/10179/294

Browse

Has files: YesSubject: search.filters.subject.460502 Data mining and knowledge discovery

Search Results

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Learning analytics : on effectiveness of dashboarding for enhancing student learning : a thesis with publications presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Information Technology, School of Mathematical & Computational Sciences, Massey University, Auckland, New Zealand
    (Massey University, 2022) Ramaswami, Gomathy
    Ongoing advancements in learning analytics have provided institutions with immense opportunities to identify and discern student learning patterns across different course offerings. These patterns can help identify those students who may be at some risk of course failure (or of course completion) as soon as possible, which further allows institutions in timely offering them guidance and support for overcoming their learning difficulties. Learning Analytics Dashboards (LAD) are currently used to deliver graphical representations of data-driven insights timeously to support management teams, instructors, and students. LADs provide a comprehensive overview on current learning environments with much use of visualizations for displaying learning patterns that can capture various aspects of the student learning experience. Hence, LADs are increasingly being used as a pedagogical approach for motivating students and supporting them in meeting their learning goals. This research study has developed a student-facing LAD that shows a snapshot of students' online learning behaviors by implementing descriptive analytics components and also incorporates machine learning in a way that enables both predictive and prescriptive analytics. The study is divided into two parts. First, a generic predictive model has been developed to identify the at-risk students across a wide variety of courses. After generating the predictive model, model explainability using anchors has demonstrated the reasoning behind predictive models to enable transparency of the predictive models and increase students’ trust as they interact with the LAD. Machine learning models used in this study have implemented prescriptive components by prioritizing which changes in learning behaviors and which learning strategies adopted by a student will most likely translate into favorable results. Second, a LAD is developed. The dashboard provides visualizations that incorporates graphical and statistical information of online behavioral student patterns as they engage with the coursework. An online student survey that gauged LAD effectiveness for its usefulness and the motivational impact of its prescriptive output to better engage students with the coursework has shown promising results. The LAD design, as far as we know, is the first in the learning analytics domain that has combined all three analytics, namely descriptive, predictive, and prescriptive. This thesis has investigated an active area of research and has paved the way for more meaningful LAD design and implementation, thereby contributing to both theory and practice.
  • Thumbnail Image
    Item
    Implementation of intelligent process automation (IPA) based clinical decision support system for early detection and screening of diabetes : this thesis is presented in partial fulfilment of the requirements for the degree of Master of Information Sciences in Information Technology, School of Natural and Computational Sciences at Massey University Albany, Auckland, New Zealand
    (Massey University, 2021) Famurewa, Oluwaseun Emmanuel
    Diabetes mellitus has become a leading cause of disease-related deaths in the world. Once an individual is diagnosed with diabetes, a series of processes will be required to keep the blood sugar regular and help avoid hyperglycemia and hypoglycemia. Self-Management of diabetes is complex and involves constant glucose monitoring, diet management, care, support, exercise, and insulin management. These processes are expensive because they require detailed record-keeping of medications, activities, and a timely report to doctors to assist them in making an informed decision that will subsequently help the patient heal. Other challenges include the high cost of treatment, lifestyle changes, education, lack of medication adherence, and treatment plans. Our approach is to adopt the Early screening technique and detect the risk of diabetes unobtrusively. Early screening is a technique that can help detect Type 1, 2 diabetes and achieve preventive care according to the guidelines set by WHO and recommended by the American Diabetes Association (ADA). Unobtrusive systems allow a doctor to screen for diabetes while he is unaware. We followed the Design Science Research model (DSRM) and started by using systematic literature review (SLR) guidelines to search the most popular journals limiting the results tied to studies that discussed the screening and detection of the risk of diabetes. We reviewed the architecture, features, and limitations of the various tools and technologies using the following classification: Continuous Glucose Monitoring Systems (CGMS), Flash Glucose Monitoring Systems (FGMS), and the Unobtrusive Systems. In addition, under the unobtrusive system, we studied the Child Health Improvement through Computer Automation (CHICA) system. While there is evidence that supports its benefits and usefulness, we found some required enhancements from the literature in the areas of decision support systems, data entry automation, and flexible integration with other systems. The artefact built during the development phase is an Intelligent process automation (IPA) system that can be implemented within the health sector for early screening and detection of diabetes unobtrusively. Developing this artefact will allow us to understand the possible issues and challenges of implementing an automation process in a medical institution. We evaluated the artefact using a mix of quantitative and qualitative methods. This method allowed us to answer the research questions and understand the value of automation to medical practitioners. The value includes speed, reduce cost, and error while safeguarding the lives of the medical professional on active duty. The results show that the system can enhance patient-doctor interaction, reduce patient wait time, and optimize the glucose monitoring process. However, there were challenges such as cost of implementation, training of staff, and the increased workload within the system. In addition, potential challenges identified include fear of job loss and aversion to change during implementation within the hospital. This study has also allowed us to understand the integration of robotic process automation with machine learning within the healthcare sector. We hope that this study will contextually position IPA within the technological stack of health care institutions and add to the body of knowledge on this subject.
  • Thumbnail Image
    Item
    Prediction of students' performance through data mining : a thesis presented in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Computer Science, Massey University, Auckland, New Zealand
    (Massey University, 2020) Umer Baloch, Rahila
    Government funding to higher education providers is based upon graduate completions rather than on student enrollments. Therefore, unfinished degrees or delayed degree completions are major concerns for higher education providers since these problems impact their long-term financial security and overall cost-effectiveness. Therefore, providers need to develop strategies for improving the quality of their education to ensure increased enrollment and retention rates. This study uses predictive modeling techniques for assisting providers with real-time identification of struggling students in order to improve their course retention rates. Predictive models utilizing student demographic and other behavioral data gathered from an institutional learning platform have been developed to predict whether a student should be classed as at-risk of failing a course or not. Identification of at-risk students will help instructors take proactive measures, such as offering students extra help and other timely supports. The outcomes of this study will, therefore, provide a safety net for students as well as education providers in improving student engagement and retention rates. The computational approaches adopted in this study include machine learning techniques in combination with educational process mining methods. Results show that multi-purpose predictive models that were designed to operate across a variety of different courses could not be generalized due to the complexity and diversity of the courses. Instead, a meta-learning approach for recommending the best classification algorithms for predicting students’ performance is demonstrated. The study reveals how process-unaware learning platforms that do not accurately reflect ongoing learner interactions can enable the discovery of student learning practices. It holds value in reconsidering predictive modeling techniques by supplementing the analysis with contextually-relevant process models that can be extracted from stand-alone activities of process-unaware learning platforms. This provides a prescriptive approach for conducting empirical research on predictive modeling with educational data sets. The study contributes to the fields of learning analytics and education process mining by providing a distinctive use of predictive modeling techniques that can be effectively applied to real-world data sets.
  • Thumbnail Image
    Item
    Mining complex trees for hidden fruit : a graph–based computational solution to detect latent criminal networks : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Information Technology at Massey University, Albany, New Zealand.
    (Massey University, 2019) Robinson, David
    The detection of crime is a complex and difficult endeavour. Public and private organisations – focusing on law enforcement, intelligence, and compliance – commonly apply the rational isolated actor approach premised on observability and materiality. This is manifested largely as conducting entity-level risk management sourcing ‘leads’ from reactive covert human intelligence sources and/or proactive sources by applying simple rules-based models. Focusing on discrete observable and material actors simply ignores that criminal activity exists within a complex system deriving its fundamental structural fabric from the complex interactions between actors - with those most unobservable likely to be both criminally proficient and influential. The graph-based computational solution developed to detect latent criminal networks is a response to the inadequacy of the rational isolated actor approach that ignores the connectedness and complexity of criminality. The core computational solution, written in the R language, consists of novel entity resolution, link discovery, and knowledge discovery technology. Entity resolution enables the fusion of multiple datasets with high accuracy (mean F-measure of 0.986 versus competitors 0.872), generating a graph-based expressive view of the problem. Link discovery is comprised of link prediction and link inference, enabling the high-performance detection (accuracy of ~0.8 versus relevant published models ~0.45) of unobserved relationships such as identity fraud. Knowledge discovery uses the fused graph generated and applies the “GraphExtract” algorithm to create a set of subgraphs representing latent functional criminal groups, and a mesoscopic graph representing how this set of criminal groups are interconnected. Latent knowledge is generated from a range of metrics including the “Super-broker” metric and attitude prediction. The computational solution has been evaluated on a range of datasets that mimic an applied setting, demonstrating a scalable (tested on ~18 million node graphs) and performant (~33 hours runtime on a non-distributed platform) solution that successfully detects relevant latent functional criminal groups in around 90% of cases sampled and enables the contextual understanding of the broader criminal system through the mesoscopic graph and associated metadata. The augmented data assets generated provide a multi-perspective systems view of criminal activity that enable advanced informed decision making across the microscopic mesoscopic macroscopic spectrum.